Power Supply latch-off
Hi,
I have a question about the failure of a home appliance power supply that is built with an IC from TinySwitch family.
I have made a SMPS for the main board of an oven with TNY277.
Circuit specs: Input 175~275 Vac 50Hz , Single output 12 V and 1 A.
The power supply works properly in the Lab and passes all standard switching tests of IEC and EMI tests(EFT, burst, DIP, Surge, etc); But when the main board is assembled on product(oven) in mass production line and tested, the SMPS fails to keep the output and the product turns off.
mass production line test conditions: The product is connected to a conventional outlet(220V, 50Hz) and turns on and off every 10 minutes (10 mins on & 10 mins off), but it suddenly turns off and remains of until next restart(disconnection & again connection to the outlet). I don't know what happens but a similar condition is explained in IC datasheet (page2, part 3 of BP/M pin explanation) that when current flowing through BP/M pin exceeds 5.5mA, The IC latches off until the voltage on this pin goes below 4.9V. When the power supply is working properly, BP/M pin voltage is 6.35V and when it fails, BP/M pin voltage is 5.85V and output voltage is 0.9V. we have tested the circuit in different input voltage conditions ( high or low voltage and distortions in wave shape) and also various load conditions ( Dynamic load, Over current, etc) in Electronics Lab but the failure doesn't happen. Failure happens frequently in mass production line.
Also measuring BP/M pin current in the Lab, it never goes beyond 3mA.
I'm thankful for helping me to find the reason of failure and proper solution for it.
Schematic of main board is attached.
Best regards.
댓글
Hi,
Could you send me the layout as well? Also, do you have waveforms of the drain-source voltage and current and BP voltage when this phenomenon occurs?
Regards,
Hi
Sorry for being late to answer.
The layout of switching power supply section is attached.
As I said before, the fault condition happens randomly and I don't have drain-source current and voltage wave shapes.
The BP voltage is 6.35V when PSU is working properly and 5.8V when it goes to auto-restart mode.
Also when I touch pin #1 (EN/UV) with a volt-meter or oscilloscope probe, the PSU starts working again; the same as Mohamed
| 첨부 파일 | 파일 크기 |
|---|---|
| Oven_SMPS_layout1.png (165.86 KB) | 165.86 KB |
| Oven_SMPS_layout2.png (164.4 KB) | 164.4 KB |
Can you try removing C02 (1nF/50V/0805)?
Introducing a capacitor in between the EN/UV pin and the optocoupler feedback may cause a slower transient response of the system. This may be the reason why you are having issues with dynamic loading.
Regards,
Hi,
Thanks Kobe for the answer.
The fault happens in no tests in Electronics Lab and testing in mass production line needs time.
Can you tell me what happens in the IC when transient response is slow; Which may cause the fault?
Which mode is activated in the IC?
As an approach for testing your opinion about the fault, I replaced C02 with bigger capacitors; first time a 100nF cap. and next time a 1uF one; But even testing the PSU in full load or dynamic load condition with new capacitors, I couldn't see the fault. Is this way reasonable for testing your opinion?
If yes, is there any other way to make the fault happen?
Regards.
Hi,
If a capacitor is connected in between the EN/UV pin of the TinySwitch and the optocoupler, the capacitor may act as a filter which would hinder accurate feedback information from getting transmitted from the output side to the controller. Instead of replacing the capacitor, I suggest you remove it completely and see how it performs in the assembly/mass production line.
Regards,
Hi Kobe,
We tested removing C02 in mass production line with 100 products, But it didn't work and the fault is yet seen.
If you have any other idea I'm glad to know about it.
By the way, replacing C03(4.7uF/50V/1206) with a 100nF/50V cap. lessens fault happening, But I don't know how it does.
Thanks for your attention.
Hi,
Based on your main schematic, you are using a 4.7uF capacitor for the BP capacitor. According to the datasheet, you may try using one of the recommended BP capacitors:
1. 0.1uF (Standard current limit)
2. 1uF (Reduced current limit)
3. 10uF (Increased current limit)
You may opt using the 10uF BP capacitor so that maximum power is delivered per cycle due to the increased current limit.
The original 4.7uF used is in between the recommended 1uF and 10uF, hence, the IC may not be delivering the needed full power at dynamic load. The current limit may be either reduced or increased because of the value of the BP capacitor used. When you used a 0.1uF (100nF) BP cap, the IC used the standard current limit, hence, the fault occurence is lessened.
To ensure maximum power delivery, you may opt to use a 10uF BP capacitor in debugging.
Regards,
I have explained mass production line test conditions wrongly.
mass production line test conditions : The oven is programmed with a test plan in which the power supply becomes full load and no load periodically every 10 seconds( 10 seconds full load & 10 seconds no load). On average from every 10 products, 2 ones suddenly get into failure & turn off until the load is removed or the input voltage is disconnected & again connected.
It is weird for us that why it happens.